Enhanced antitumor potential induced by chloroacetate-loaded benzophenones acting as fused tubulin-pyruvate dehydrogenase kinase 1 (PDHK1) ligands

Article abstract of DOI:10.1016/j.bioorg.2020.103643

The majority of cancers detected every year are treated with anti-cancer compounds. Unfortunately, many tumors become resistant to antineoplastic drugs. One option is to use cocktails of compounds acting on different targets to try to overcome the resista

Full text of DOI:10.1016/j.bioorg.2020.103643

Journal Pre-proofs
Enhanced antitumor potential induced by chloroacetate-loaded benzophe-
nones acting as fused tubulin-pyruvate dehydrogenase kinase 1 (PDHK1) li-
gands
Alina Ghinet, Xavier Thuru, Emilie Floquet, Joëlle Dubois, Amaury Farce,
Benoît Rigo
PII:
S0045-2068(19)31937-6
DOI:
https://doi.org/10.1016/j.bioorg.2020.103643
YBIOO 103643
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
13 November 2019
23 January 2020
29 January 2020
Please cite this article as: A. Ghinet, X. Thuru, E. Floquet, J. Dubois, A. Farce, B. Rigo, Enhanced antitumor
potential induced by chloroacetate-loaded benzophenones acting as fused tubulin-pyruvate dehydrogenase kinase
1 (PDHK1) ligands, Bioorganic Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.2020.103643
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Inc.

Enhanced antitumor potential induced by chloroacetate-loaded benzophenones
acting as fused tubulin-pyruvate dehydrogenase kinase 1 (PDHK1) ligands
Alina Ghinet,a,b,c,* Xavier Thuru,d Emilie Floquet,d,e Joëlle Dubois,f Amaury Farce,g Benoît Rigoa,b
a Yncréa Hauts-de-France, Hautes Etudes d’Ingénieur (HEI), UCLille, Health & Environment Department,
Team Sustainable Chemistry, Laboratoire de Chimie Durable et Santé, 13 rue de Toul, F-59046 Lille,
France
b
Univ. Lille, Inserm, CHU Lille, Institut Pasteur de Lille, U1167 - RID-AGE - Facteurs de risque et
déterminants moléculaires des maladies liées au vieillissement, F-59000 Lille, France
c ‘Alexandru Ioan Cuza’ University of Iasi, Faculty of Chemistry, Bd. Carol I nr. 11, 700506 Iasi, Romania
d
Univ. Lille, CNRS, Inserm, CHU Lille, UMR9020 – UMR1277 - Canther – Cancer Heterogeneity,
Plasticity and Resistance to Therapies, F-59000 Lille, France
e Plateau de Cytométrie, BioImaging Center Lille Nord de France, BICeL Campus Hospitalo-Universitaire,
Lille, France
f Institut de Chimie des Substances Naturelles, UPR2301 CNRS, Centre de Recherche de Gif, Avenue de la
Terrasse, F-91198 Gif-sur-Yvette Cedex, France
g Université de Lille, Faculté des Sciences Pharmaceutiques et Biologiques, Rue du Pr Laguesse, B.P. 83,
F-59006 Lille, France
Corresponding author: alina.ghinet@yncrea.fr
Abstract
The majority of cancers detected every year are treated with anti-cancer compounds. Unfortunately,
many tumors become resistant to antineoplastic drugs. One option is to use cocktails of compounds
acting on different targets to try to overcome the resistant cells. This type of approach can produce good
results, but is often accompanied by a sharp increase of associated side effects.
The strategy presented herein focuses on the use of a single compound acting on two different biological
targets enhancing potency and lowering the toxicity of the chemotherapy. In this light, the approach
presented in the current study involves the dual inhibition of human pyruvate dehydrogenase kinase-1
(PDHK1) and tubulin polymerization using mono-, di- and tri-chloroacetate-loaded benzophenones and
benzothiophenones.
1

Synthesized molecules were evaluated in vitro on tubulin polymerization and on pyruvate
dehydrogenase kinase 1. The cell cycle distribution after treatment of DA1-3b leukemic cells with active
compounds was tested.
Twenty-two benzo(thio)phenones have been selected by the National Cancer Institute (USA) for
evaluation of their anti-proliferative potential against NCI-60 cancer cell lines including multidrug-
resistant tumor cell lines. Seventeen molecules proved to be very effective in combating the growth of
tumor cells exhibiting inhibitory activities up to nanomolar range.
The molecular docking of best antitumor molecules in the study was realized with GOLD in the tubulin
and PDHK1 binding sites, and allowed to understand the positioning of active molecules.
Chloroacetate-loaded benzo(thio)phenones are dual targeted tubulin- and pyruvate dehydrogenase
kinase 1 (PDHK1)- binding antitumor agents and exhibited superior antitumor activity compared to non-
chlorinated congeners particularly on leukemia, colon, melanoma and breast cancer cell lines.
Keywords
DCA, dichloroacetic acid; PDHK1 (h), human pyruvate dehydrogenase kinase 1; SAR, structure-activity
relationship; benzophenone; chloroacetate.
1. Introduction
Cancer remains the second leading cause of death in the world and is expected to increase over 70% in
the next two decades according to WHO [1]. Nearly one in six deaths worldwide is caused by cancer.
The majority of the new cancer cases detected every year is treated with anti-cancer compounds.
Unfortunately, many tumors become resistant to these antineoplastic drugs. One option is to use
cocktails of compounds acting on different targets to try to overcome the resistant cells. This type of
approach can produce good results, but is often accompanied by a sharp increase of associated side
effects.
Another strategy is to use a single compound acting on two different biological targets enhancing
potency and lowering the toxicity of the chemotherapy [2]. In this light, the innovative approach
presented in the current study involves the simultaneously inhibition of PDHK-1 (h) and tubulin. To the
best of our knowledge, this is the very first therapeutic proposal using dual inhibitors acting on these
two particular targets involved in cancer development.
The first point of action is the inhibition of the tubulin. This protein is a largely studied biological target
in oncology, involved in cell proliferation for its ability to be polymerized and form microtubules, key
components of the cytoskeleton. Tubulin is the target of many small molecules that interfere with the
2

dynamics of polymerization or depolymerization. Most of them bind to the taxol, vinblastine or
colchicine binding site. Several tubulin inhibitors are already marketed (e. g. colchicine, paclitaxel,
vinflunine, vincristine, Halichondrin B, etc.) or in clinical development (e. g. combretastatin A-4
derivatives, 2-methoxyestradiol, etc) [3].
Dichloroacetic acid (DCA) and sodium dichloroacetate have been used in human therapy for over thirty
years especially in the treatment of mitochondrial disorders. DCA is a non-specific inhibitor of the
mitochondrial pyruvate dehydrogenase kinases (PDK1-4 or PDHK1-4) which thus activates pyruvate
dehydrogenase complex (PDH) that manage the control between oxidative phosphorylation and
glycolysis. The PDH is located in the mitochondrial matrix catalyzing the irreversible decarboxylation of
pyruvate to acetyl-CoA and NADH [4].
DCA is currently being investigated as a potential target in oncology and studied in different clinical trials.
A completed phase 1 study evaluated the safety and tolerability of oral DCA in the treatment of recurrent
malignant brain tumors (RMBTs). DCA was also used in combination with Cisplatin and radiation
treatment in patients with stage III-IV squamous cell carcinomas of the head and neck (SCCHN) in a phase
2 clinical trial regrouping fifty subjects. Another active clinical trials explore the benefit of DCA in the
vascular remodeling in PAH (early phase 1), in patients with breast and non-small cell lung cancer (phase
2), diabetic hyperglycaemia (phase 1) or in congenital lactic acidosis (phase 3) [3].
Increasing interest is currently being paid for DCA analogues. Multiple DCA-loaded amides [5] and
dichloroacetophenones [6] have been recently developed as potent PDHK targeting compounds and
displayed cell growth inhibition on leukemic cells [5] and non-small cell lung cancer cells [6].
The poor pharmacokinetic parameters of DCA (low lipophilia, low permeability through cell membranes
and short half-time life impose high dosages to achieve therapeutic efficiency) prevent its use for cancer
treatment. To overcome this inconvenient and in order to build strong SAR in these series, we have
designed and synthesized novel DCA-carrier molecules by associating anti-tubulin benzophenones
derived from phenstatin, known potent tubulin polymerization inhibitor, with mono-, di- and tri-
chloroacetate groups targeting PDHK1 (Figure 1). Moreover, these hybrid molecules may constitute
mutual prodrugs (codrugs). Through hydrolysis in the tumor cell biological system and hydrolytic
cleavage promoted by hepatic microsomal esterases or amidases, these potential codrugs can provide
both anti-cancer compounds (e.g. phenstatin and DCA), without releasing other molecular entities with
unknown mode of action and ADMET profile.
3

O
O
O
O
OH
O
Cl
A
B
HO
Cl
O
Dichloroacetic acid (DCA)
PDHK1-4 inhibitor
Phenstatin
tubulin polymerization inhibitor
OH, OCH3, OCOCH2Cl, OCOCHCl2
S
F
F
Cl
Cl
Cl
Cl
O
O
F
O
O
O
Cl
Cl
O
O
O
O
O
O
O
Cl
O
R
N
R
N
R
Cl
O
O
N
O
Cl
O
O
O
target dual tubulin/PDHK1 inhibitor
O
Figure 1. Structure of phenstatin, dichloroacetic acid (DCA) and of target dual inhibitors decorated with
acetyl and mono-, di- and trichloroacetyl units
2. Results and discussion
2.1. Chemistry
All benzophenones described in the current manuscript, and also used as starting material for the
synthesis of thioketones 4a-b, were obtained by Eaton's reagent mediated Friedel-Crafts reaction. This
procedure has already been described for ketones 3a,c,e,k,m,o,p,r,t (Scheme 1) [7-8]. During these
syntheses, the phenol groups were protected as monochloroacetic ester (reagents 20 and 21, Scheme
1) in order to avoid side reactions as previously observed when using acetate protective group for
phenols [9], while the amino group was protected as acetamide function (reagents 22 and 23, Scheme
1). In the sole case of the Friedel-Crafts acylation promoted by Eaton’s reagent between the 2,4,5-
trifluoro-3-methoxybenzoic acid 17 and 2,6-dimethoxyphenyl chloroacetate 21, the structure of the final
benzophenone isolated after flash liquid chromatography purification of the crude on silica column did
not correspond to the expected one. A deprotection of the ester function, and a mono O-demethylation
occurred, leading to benzophenone 3o as the unique reaction product. The O-demethylated position of
this phenol was secured by NOESY study (Scheme 1).
In order to provide an additional chemical modulation on the A ring of parent phenstatin, the classical
3,4,5-trimethoxyphenyl unit (ring A, Figure 1) was replaced by 3,5-dimethyl-4-methoxyphenyl unit by
using the same synthetic strategy to yield benzophenones 3g and 3i. Chloroacetyl protected ketones
3a,c,g,i,m,p,r,t were next reacted with sodium acetate in refluxing methanol to yield ketones with free
phenol group 3b,d,h,j,n,q,s,u while acetamide functions of products 3e and 3k were deprotected in
4

acidic conditions in refluxing methanol to provide free amines 3f and 3l. Next, the thionation of
phenstatin 3b and its monochloroacetic ester analogue 3a with the reagent combination of P4S10 and
hexamethyldisiloxane [10] provided the first synthesis of thiophenstatin 4b and its monochloroacetic
derivative 4a in 58 and 63% yields, respectively (Scheme 1).
3a, R1 = R5 = H, R2 = R3 = R4 = R7 = OMe, R6 = OCOCH2Cl, 85%
(ii)
3b, R1 = R5 = H, R2 = R3 = R4 = R7 = OMe, R6 = OH, 98%
3c, R1 = H, R2 = R3 = R4 = R5 = R7 = OMe, R6 = OCOCH2Cl, 80%
3d, R1 = H, R2 = R3 = R4 = R5 = R7 = OMe, R6 = OH, 62%
3e, R1 = R5 = H, R2 = R3 = R4 = R7 = OMe, R6 = NHCOCH3, 53%
3f, R1 = R5 = H, R2 = R3 = R4 = R7 = OMe, R6 = NH2, 87%
3g, R1 = R5 = H, R2 = R4 = Me, R3 = R7 = OMe, R6 = OCOCH2Cl, 80%
3h, R1 = R5 = H, R2 = R4 = Me, R3 = R7 = OMe, R6 = OH, 92%
3i, R1 = H, R2 = R4 = Me, R3 = R5 = R7 = OMe, R6 = OCOCH2Cl, 87%
3j, R1 = H, R2 = R4 = Me, R3 = R5 = R7 = OMe, R6 = OH, 98%
3k, R1 = H, R2 = R4 = Me, R3 = R5 = R7 = OMe, R = NHCOCH3, 66%
3l, R1 = H, R2 = R4 = Me, R3 = R5 = R7 = OMe, R = NH2, 85%
3m, R1 = R3 = R4 = F, R2 = R7 = OMe, R5 = H, R6 = OCOCH2Cl, 90%
3n, R1 = R3 = R4 = F, R2 = R7 = OMe, R5 = H, R6 = OH, 96%
3o, R1 = R3 = R4 = F, R2 = R7 = OMe, R5 = R6 = OH, 71%
R
R
5
R
O
R
5
1
1
R
CO2H
R
6
(ii)
R
R
6
2
(i)
2
+
(iii)
R
R
7
R
R
7
3
3
R
R
4
4
(ii)
(ii)
3a-u
20-23
15-19
20, R5 = H, R6 = OCOCH2Cl, R7 = OMe
21, R5 = R7 = OMe, R6 = OCOCH2Cl
22, R5 = R7 = OMe, R6 = NHCOCH3
23, R5 = H, R6 = NHCOCH3, R7 = OMe
15, R1 = H, R2 = R3 = R4 = OMe
(iv)
16, R1 = H, R2 = R4 = Me, R3 =OMe
17, R1 = R3 = R4 = F, R2 = OMe
(iii)
(ii)
18, R1 = H, R2 = R4 =OMe, R3 = OCO2Et
19, R1 = H, R2 = R4 = OMe, R3 = Me
3p, R1 = R5 = H, R2 = R4 = R7 = OMe, R3 = OCO2Et, R6 = OCOCH2Cl, 90%
3q, R1 = R5 = H, R2 = R4 = R7 = OMe, R3 = OCO2Et, R6 = OH, 93%
3r, R1 = H, R2 = R4 = R5 = R7 = OMe, R3 = OCO2Et, R6 = OCOCH2Cl, 82%
3s, R1 = H, R2 = R4 = R5 = R7 = OMe, R3 = OCO2Et, R6 = OH, 90%
3t, R1 = R5 = H, R3 = Me, R2 = R4 = R7 = OMe, R6 = OCOCH2Cl, 78%
3u, R1 = R5 = H, R3 = Me, R2 = R4 = R7 = OMe, R6 = OH, 95%
(ii)
S
O
O
O
O
R
(ii)
(ii)
8
4a, R8 = OCOCH2Cl, 63%
4b, R8 =H, 58%
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
OH
O
O
O
OH
O
Cl
Cl
O
O
O
O
O
O
O
3d, 62%
3b, 98%
Phenstatin
3a, 85%
3c, 80%
O
O
O
O
H
N
NH2
O
O
O
O
O
O
OH
O
Cl
O
O
O
O
O
O
O
3g, 80%
3e, 53%
3f, 87%
3h, 92%
O
O
O
O
O
O
O
O
H
N
NH2
O
O
OH
O
Cl
O
O
O
O
O
O
O
3i, 87%
3j, 98%
3l, 85%
3k, 66%
O
F
O
F
F
O
F
O
OH
O
O
O
O
F
O
O
O
F
OH
O
O
F
OH
Cl
Cl
O
O
O
O
O
O
O
F
F
3p, 90%
3o, 71%
3n, 96%
3m, 90%
O
O
O
O
O
O
O
OH
O
O
O
OH
O
O
O
O
O
O
O
Cl
O
Cl
O
O
O
O
O
O
O
O
O
O
O
O
O
3t, 78%
3s, 90%
3r, 82%
3q, 93%
O
S
S
O
OH
O
O
O
O
O
OH
Cl
O
O
O
O
O
O
O
3u, 95%
4a, 63%
4b, 58%
Thiophenstatin
Scheme 1. Synthesis of monochloroacetate carriers, acetamides 3e and 3k and corresponding free phenols
and amines. Reagents and conditions: (i) 4 equiv. Eaton’s reagent, 60°C, 1-24h; (ii) 4.5 equiv.
5

CH3COONa.3H2O, MeOH, reflux, 2-10 h; (iii) 1.5 equiv. aq HCl 10%, MeOH, reflux, 24h; (iv) 0.5 equiv.
phosphorus pentasulfide (P4S10), 2 equiv. hexamethyldisiloxane (HMDSO), toluene, reflux, 24 h. Compounds
numbered in blue were already described in the literature: benzophenones 3a-e and 3p-s have been reported
in ref. [7]; benzophenone 3f was reported in ref. [11]; benzophenones 3m and 3n have been reported in ref.
[8]; benzophenone 3t has been reported in ref. [12].
In order to perform the different modulations allowing to enrich structure-activity relationships in this series,
and to conclude on the importance of the chloroacetic ester unit, the replacement by an acetic ester was
first envisaged by synthesizing acetylphenstatin 3x in 87% yield upon treatment of phenstatin 3b with acetic
acid chloride at reflux for 1 hour (Scheme 2). The bisphenol 3o was reacted with 6 equivalents of
monochloroacetic acid chloride to obtain a bis-monochloroacetate carrier which possibly could improve the
activity against human PDHK1. The reaction provided the target diester 3v in 41% yield as the major product
along with the monoester 3w in similar yield (Scheme 2). Next, the treatment of free amine 3l with 4
equivalents of monochloroacetic acid chloride at reflux for 4 hours resulted in the synthesis of bis-
1
chloroacetyl amide carrier 5 in excellent 98% yield (Scheme 2). The H NMR monitoring of the reaction
progress indicated the presence of the diacylated product 5 very quickly thus preventing the isolation of the
monoacylated product in these operatory conditions.
R
O
R
5
R
O
R
5
1
1
R
O
O
R
OH
O
(i)
R
2
9
2
O
R
R
3
3
R
R
4
4
3v-x
3b, 3o
3v, R1 = R3 = R4 = F, R2 = OMe, R5 = OCOCH2Cl, R9 = CH2Cl, 41%
3w, R1 = R3 = R4 = F, R2 = OMe, R5 = OH, R9 = CH2Cl, 37%
3x, R1 = R5 = H, R2 = R3 = R4 = OMe, R9 = Me, 87%
O
O
O
O
O
Cl
NH2
O
(i)
N
O
Cl
O
O
O
3l
5
Cl
F
F
O
OH
F
F
O
O
O
O
O
F
O
O
O
F
Cl
Cl
O
O
O
3w, 37%
3v, 41%
Cl
O
O
O
O
O
N
O
O
O
Cl
O
O
O
O
O
5, 98%
3x, 87%
Scheme 2. Synthesis of additional monochloroacetate carriers 3v, 3w, acetate 3x and bis-
monochloroacetamide 5. Reagents and conditions: (i) 2-6 equiv. monochloroacetic acid chloride or acetic
acid chloride, reflux, 1-4h. Benzophenone 3x numbered in blue has been previously reported in ref. [13].
Afterward, the chemical efforts turned on the same benzophenones but decorated with dichloroacetate
groups first, and then with trichloroacetate groups in a second time (Scheme 3). All these synthetic strategies
were deployed to obtain a clear trend in antitumor activity related to the nature of the chloroacetate
(monochloro, dichloro or trichloroacetic group). Phenstatin 3b and benzophenones 3h, 3j and 3o were
treated with dichloroacetic acid chloride (2 equivalents for monoacylated benzophenones 6a-c and 6
6

equivalents for diacylated benzophenone 6d) in the absence of base and provided corresponding
dichloroacetates 6a-d in good to excellent yields (63-95%) (Scheme 3). Similar synthetic strategy was applied
to easily obtain trichloroacetates 7a and 7b by reacting phenstatin 3b and benzophenone 3h with 2
equivalents of trichloroacetic acid chloride (Scheme 3). The treatment of benzophenone 3l bearing free
amino group with dichloroacetic chloride provided bisacylated derivative 8 in excellent 98% yield (Scheme
3). In the same way as during the synthesis of the bismonochloroacetylated congener 5, the monoacylated
compound has not been observed in the crude.
6a, R1 = R5 = H, R2 = R3 = R4 = OMe, R9 = CHCl2, 95%
6b, R1 = R5 = H, R2 = R4 = Me, R3 =OMe, R9 = CHCl2, 77%
6c, R1 = H, R2 = R4 = Me, R3 = R5 = OMe, R9 = CHCl2, 88%
6d, R1 = R3 = R4 = F, R2 = OMe, R5 = OCOCHCl2, R9 = CHCl2, 63%
7a, R1 = R5 = H, R2 = R3 = R4 = OMe, R9 = CCl3, 84%
7b, R1 = R5 = H, R2 = R4 = Me, R3 =OMe, R9 = CCl3, 65%
R
O
R
5
R
O
R
5
1
1
R
O
O
R
OH
O
(i)
R
2
9
2
O
R
R
3
3
R
R
4
4
6a-d, 7a-b
3b, 3h, 3j, 3o
O
CHCl2
CHCl2
O
3l
O
O
O
O
NH2
O
(i)
N
O
O
O
O
8, 80%
Cl
Cl
F
F
O
O
O
O
Cl
Cl
Cl
O
O
Cl
O
O
O
O
O
O
O
O
O
O
Cl
Cl
Cl
Cl
O
O
O
O
O
F
O
O
O
O
O
Cl
6d, 63%
6b, 77%
6c, 88%
6a, 95%
Cl
Cl
Cl
O
O
Cl
O
Cl
Cl
Cl
Cl
N
O
O
O
O
O
O
Cl
O
O
O
O
O
O
O
8, 98%
7a, 84%
7b, 98%
Scheme 3. Synthesis of dichloroacetate and trichloroacetate carriers. Reagents and conditions: (i) 2-6 equiv.
dichloroacetic acid chloride or trichloroacetic acid chloride, reflux, 1-6h.
2.2.
In Vitro Pharmacology:
2.2.1. Human pyruvate dehydrogenase kinase (PDHK1, 2 and 4) enzymatic radiometric assay [km ATP],
KinaseProfiler
The newly synthesized benzophenones and benzothiophenones and reference chloroacetic acids
(monochloroacetic acid, dichloroacetic acid and trichloroacetic acid) and sodium dichloroacetate were
submitted to biological evaluation on human PDHK1. The selectivity of reference dichloroacetic acid was
studied on PDHK1, 2 and 4. Compounds were tested at 100 and 1000 µM concentration. Compound PDHK
(h) inhibition effect was calculated as a % inhibition of control kinase activity. Results showing an inhibition
higher than 50% are considered to represent significant effects of the test compounds. The PDHK assay of
molecules was realized by Eurofins Cerep SA, Celle L'Evescault, France.
2.2.1.1.
Selectivity of DCA on PDHK1, PDHK2 and PDHK4 (h)
Many papers describe DCA as a non-specific inhibitor of the mitochondrial pyruvate dehydrogenase kinases
PDHK1-4. The selectivity of DCA was also evaluated in this study on PDHK-1, 2 and 4 via the enzymatic
radiometric assay [km ATP] KinaseProfiler. DCA was tested at 1mM concentration. It is interesting to note
that no inhibitory effect was registered on PDHK4 at this dose while similar low activities were measured
on PDHK1 and PDHK2 (14% and 16% inhibition, respectively) (Figure 2). There is a lot of confusion in the
literature concerning the DCA abbreviation. It refers to both dichloroacetic acid or sodium dichloroacetate
and sometimes to the chloroacetate anion without specifying the counter ion [14]. This confusion may
7

explain the differences in inhibitory activity among these chemical entities, which are different from one
another. For instance, the behavior of dichloroacetate on PDHK1 and PDHK2 was already reported at 0.1-1
mM concentration by Makita et al. 2017 [14]. There was no important difference in inhibitory activities for
concentrations of dichloroacetate below 1mM. However, at concentrations above 1 mM, the
dichloroacetate behavior is reversed inhibiting PDHK1 two to three more times than PDHK2 at increasing
concentrations (e.g. 10 mM) [14].
In this study, we used exclusively dichloroacetic acid as the reference molecule and abbreviated DCA.
100
PDHK4
84
PDHK2
PDHK1
86
80
85
90
95
100
Kinase activity in presence of DCA (%)
Figure 2. Selectivity of DCA on PDHK1, PDHK2 and PDHK4 (h): DCA slightly inhibits PDHK1 and PDHK2 and
has no impact on PDHK4 at 1mM concentration.
2.2.1.2.
Effects of selected molecules on human PDHK1
Fifteen mono-, di- and tri-chloroacetate carriers were tested for their ability to inhibit human PDHK1 in vitro
at two different concentrations (1 mM and 100 µM). DCA and sodium dichloroacetate were used as
reference compounds in the same assay. Monochloroacetic acid (MCA) and trichloroacetic acid (TCA) were
also included in order to compare their activity with that of known PDHK1 inhibitor DCA and to serve as
references for mono- and trichloroacetate-loaded molecules, respectively. Results are provided in Table 1.
Dichloroacetic acid exhibited enhanced inhibitory potential compared to sodium dichloroacetate. Esters 3g,
3v, 4a and 6d showed improved inhibitory potential compared to DCA. Monochloroacetate esters 3p and
3w and bis-monochloroacetamide 5 displayed equivalent activity to that of reference DCA. It is interesting
to note that the best PDHK1 inhibitor 3v is decorated by two monochloroacetate units and is much more
active than related analogue 6d bearing two dichloroacetate units. Contrary to what was expected, the
monochloroacetate constituted the best pharmacomodulation in the current study for targeting human
PDHK1 among designed mono-, di- and tri- chloroacetates. Compounds 3v and 6d were also found to
weakly inhibit tubulin polymerization (48% and 42% inhibition at 100 µM, respectively). Interestingly, the
benzophenone 3g and benzothiophenone 4a, potent tubulin polymerization inhibitors (IC50 of 1.30 µM and
1.17 µM, respectively), also exhibited additional inhibitory activity against PDHK1 (h) 2.3 times greater than
dichloroacetic acid and constituted the best potential codrugs from the series.
2.2.2. Effects of synthesized molecules on tubulin polymerization
The potential of synthesized benzophenones and benzothiophenones to inhibit tubulin polymerization in
vitro was first tested at a single concentration of 100 µM (Table 1). Phenstatin 3b and
desoxypodophyllotoxin have been used as positive references and DMSO as negative control. Molecules
that resulted in more than 50% inhibition were selected for IC50 calculation. Eighteen molecules proved to
be excellent tubulin polymerization inhibitors exhibiting IC50 values in the micromolar and submicromolar
range. The replacement of the carbonyl unit of phenstatin 3b with a thiocarbonyl (thiophenstatin 4b)
improved the antitubulin efficiency. The classical A ring (3,4,5-trimethoxyphenyl unit) of phenstatin has
been successfully replaced by a 3,5-dimethyl-4-methoxyphenyl ring in compound 3h, slightly improving the
8

inhibitory effect. On the other hand, all other modifications of the classical A ring (the replacement by 3,5-
dimethoxy-4-methylphenyl in molecule 3u, 2,4,5-trifluoro-3-methoxyphenyl in molecule 3n or 3,5-
dimethoxy-4-ethylcarbonatephenyl unit in molecule 3s) conserved or slightly diminished the antitubulin
potential compared to parent phenstatin.
Table 1. Evaluation of synthesized molecules on tubulin polymerization and on pyruvate dehydrogenase kinase
1 in vitro
Molecule
% TPIa,b
IC50 (tubulin) R2c
(µM)b
% PDHK1 (h)e
1 mM
0.1 mM
3a
98
91
98
0
1.21
1.34
1.22
0.9299
-
0
3c
0.9264
3d
0.9625
d
3e
-
-
3g
95
93
96
0
1.30
0.96
0.69
-
0.8871
32
0
3h
0.8702
3j
0.8920
3k
-
3l
95
5
1.69
-
0.9208
3m
-
3n
88
57
97
97
72
88
48
32
89
96
100
18
97
-
28.23
-
0.9676
3p
-
15
-
-
3r
0.88
1.60
5.30
5.07
-
0.9415
1.2
3s
0.9355
3t
0.9831
3u
0.9410
3v
-
93
15
-
55
-
3w
-
-
3x
5.99
1.17
0.75
-
0.9802
0
6
-
4a
0.9312
33
-
Thiophenstatin 4b
0.9259
5
-
12
0
2
0
0
0
10
0
0
-
6a
6b
6c
6d
7a
8
1.17
-
0.9393
-
-
94
42
93
3
0.26
-
0.9103
-
-
78
0
0.79
-
0.9272
-
0
Phenstatin 3b
96
100
-
3.42
1.76
-
0.9378
-
Desoxypodophyllotoxin
Monochloroacetic acid (MCA)
Dichloroacetic acid (DCA)
Sodium dichloroacetate
Trichloroacetic acid (TCA)
0.9740
-
-
-
-
-
-
12
14
3
-
0
-
9
-
0
-
-
-
16
-
a Inhibition of tubulin polymerization at a 100 µM concentration
b Values represent mean of two experiments
9

c Standard deviation
d Not determined
e Inhibition of human pyruvate dehydrogenase kinase 1 (PDHK1); values represent mean of two experiments
2.2.3. Cell cycle analysis on DA1-3b cells
Cell cycle analyses of newly synthesized chloroacetyl-thiophenstatin 4a and thiophenstatin 4b were
performed by flow cytometry (Figure 3). The two thiophenstatins acted differently on cell cycle arrest.
Thiophenstatin 4b strongly induced G2/M arrest in DA1-3b cells [15]. The effect was observed in a dose-
independent manner after treatment for 24 h with increased concentrations of this compound (Figure 3).
These results confirmed that thiophenstatin is a potent antitubulin agent and acts as a classic phenstatin
on the cell cycle. Treated murine leukemia DA1-3b cells showed decreased G1 population (22.9%)
compared with 53.8% in the control when treated with 1 µM of chloroacetyl-thiophenstatin 4a (Figure 3).
Whereas, the percentages of sub-G1 phase (apoptotic cells) were significantly increased after cells were
treated with 1, 10 and 100 µM of 4a up to 41.8% respectively compared with 0.7% in the control (Figure 3).
This may be related to the dual inhibition of both tubulin polymerization and PDHK1 which induces
increased apoptotic effect. The reference DCA had no effect on DA1-3b cells at low concentration (10 nM
to 10 mM) and was highly cytotoxic at high concentration (50 mM, 100 mM or 150 mM) (see supplementary
section for full results).
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
SubG1
0.7
22.2
41.8
36.7
1.0
28.1
21.1
13.4
G1
S
G2/M
17.5
18.1
18.3
21.7
17.9
27.6
39.1
57.0
>4N
1.1
1.0
1.2
1.5
1.0
0.5
0.8
0.6
(a) DMSO
53.8
29.1
20.6
22.9
54.8
32.9
24.4
13.9
26.9
29.6
18.1
17.2
25.3
10.9
14.5
15.1
(b) 4a 100 µM
(c) 4a 10 µM
(d) 4a 1 µM
(e) DMSO
(f) 4b 100 µM
(g) 4b 10 µM
(h) 4b 1 µM
Figure 3. Cell cycle distribution after treatment of DA1-3b cells with chloroacetyl-thiophenstatin 4a and
thiophenstatin 4b. DA1-3b cells were incubated for 24h with DMSO (control) and chloroacetyl-thiophenstatin 4a
10

or thiophenstatin 4b. Cell cycle parameters were analyzed by flow cytometry, as described in Methods. Values
(lower table) are expressed as percentage of the total cell population.
2.2.4. Effects of synthesized molecules on cancer cell growth
Twenty-two synthesized benzo(thio)phenones 3a, 3d, 3g-h, 3j-o, 3r-t, 3v, 3x, 4a, 5, 6a, 6c, 6d, 7a and 8 have
been selected by the National Cancer Institute (NCI) for evaluation of their anti-proliferative potential
against the NCI-60 cancer cell lines, including multidrug-resistant (MDR) tumor cell lines (HCT-15: colorectal
adenocarcinoma; NCI/ADR-RES: human ovary adenocarcinoma; RXF 393: human kidney poorly
differentiated hypernephroma; MCF7: human breast adenocarcinoma and SF-539: human CNS
glioblastoma). Resumed representative biological efficacy is described in Table 2. All selected molecules
underwent first the NCI 60 cell one-dose screen at 10 µM concentration. Only compounds which satisfied
pre-determined threshold inhibition criteria in a minimum number of cell lines have progressed to the full
5-dose assay. Seventeen molecules exhibited significant cell growth inhibition in the one-dose screen and
were evaluated against the 60 cell panel at five concentration levels (100 µM, 10 µM, 1 µM, 100 nM and 10
nM) in order to calculate the GI50 value on each cell line (Table 2). The molecules proved to be very effective
in combating the growth of tumor cells exhibiting average activities over all NCI-60 cancer cell lines (MG-
MID values) ranging from 2884 to 39 nM (Table 2).
The monohloroacetylated phenstatin 3a displayed the most important anti-proliferative activity on
leukemia, colon, melanoma and SNC cancer cells (GI50 < 10 nM). Interestingly, the thio analogue 4a of
benzophenone 3a exhibited improved anti-proliferative activity with reinforced inhibition particularly on
non-small cell lung cancer cells, A498 renal cancer and MCF-7 breast cancer cells (Table 2). The replacement
of the monochloroacetate in compound 3a by a trichloroacetate unit in compound 7a was tolerated and
provided excellent cell growth inhibition predominantly against leukemic, colon cancer, melanoma, ovarian
and CNS cancer cells (GI50 < 10 nM). The acetylated benzophenone 3x displayed similar potential to
monochloroacetylated 3a but diminished activity compared to trichloroacetylated phenstatin 7a (compare
3x to 7a, Table 2). The replacement of the classical 3,4,5-trimethoxyphenyl unit of compound 3a by a 3,5-
dimethyl-4-methoxyphenyl unit in compound 3g constituted a well-tolerated pharmacomodulation
providing the most cytotoxic agent in the study (GI50 < 10 nM on the most part of tested cell lines) (Table
2). The suppression of the monochloroacetate group of 3g in benzophenone 3h resulted in a small decrease
of the anti-proliferative activity and suppressed the activity on some cell strains (compare 3g to 3h, Table
2), underlining the importance of the chloroacetate unit. The addition of methoxy unit in the 2’ position on
the B-ring in compound 3j slightly diminished the cellular activity (compare 3h and 3j, Table 2). The same
modulation realized on phenstatin provided compound 3d that showed greater anti-proliferative potential
compared to analogue compound 3j particularly on K-562, MDR NCI/ADR-RES, HS 578T and BT-549 cell lines
(Table 2). Acetamide analogue 3k displayed diminished cell growth inhibition compared to related analogue
3l bearing a free amine group on the B ring (compare 3k and 3l, Table 2). The least cytotoxic derivative in
the NCI evaluation was benzophenone 3n bearing an original 2,4,5-trifluoro-3-methoxy phenyl unit as the
A ring displaying cell growth inhibition in the micromolar range. The benzophenone 3r decorated with a
monochloroacetate displayed equivalent potential as analogue benzophenone 3s. The replacement of the
classical A ring of benzophenone 3a by a 3,5-dimethoxy-4-methylphenyl unit in compound 3t was less
tolerated and diminished the anti-proliferative activity. This modulation was also less interesting than the
replacement of the A ring by a 3,5-dimethyl-4-methoxyphenyl unit in compound 3g.
Benzophenone 6c decorated with a dichloroacetate unit displayed high anti-melanoma potential strongly
inhibiting the MDA-MB-435 cells (GI50 < 10 nM) but also CNS cancer cells. Interestingly, the replacement of
the acetyl unit by two monochloroacetyl groups in benzophenone 5 was tolerated and improved the
antitumor activity (Table 2). The same trend was registered when changing the acetyl unit of compound 3k
by two dichloroacetyl groups in compound 8.
11

12

Table 2 Results of the in vitro human cancer cell growth inhibition for selected compounds 3a, 3d, 3g-h, 3j-l, 3n, 3r-t, 3x, 4a, 5, 6c, 7a and 8a,b
Compound
3a
3d
3g
3h
3j
3k
3l
3n
3r
3s
3t
3x
4a
5
6c
7a
8
Cell type
Cell line
GI50 (nM)a,b
c
HL-60(TB)
K-562
CCRF-CEM
<10
<10
23.5
23.5
<10
28.8
<10
<10
<10
<10
<10
<10
27.6
31.1
34.8
-
-
2020
186
3310
-
-
19.0
35.4
293
19.9
<10
38.6
<10
<10
<10
31.6
52.6
204
25.1
31.9
38.2
<10
<10
33.6
30.6
43.1
71.3
Leukemia
1090
1830
14.5
30.4
457
662
416
410
A549/ATCC
NCI-H522
NCI-H322M
363
15.8
75.1
41.3
<10
<10
40.4
<10
10.4
1200
-
-
399
35.9
59.5
284
54.8
98.8
34.8
17
42.1
4390
1340
4470
3390
264
3410
3960
311
2200
2150
217
580
55.2
24.7
209
174
10.4
47.0
97.3
56.1
67.2
330
41.7
69.9
54.7
14
93.2
94.8
48.7
57.4
Non-Small Cell Lung
Cancer
KM12
SW-620
HCT-15
<10
<10
<10
<10
24.6
10.6
<10
<10
<10
<10
<10
<10
28.9
34.7
36.5
141
167
397
19.6
32.9
26.3
1030
703
1760
543
455
415
520
489
420
90.2
123
51.1
17.7
33.8
18.4
<10
<10
<10
40.1
43.3
56.2
29.3
37.4
34.9
<10
<10
<10
40.1
43.7
58.5
Colon Cancer
Melanoma
M14
MDA-MB-435
LOX IMVI
<10
<10
<10
<10
<10
26.9
<10
<10
<10
<10
<10
<10
21.5
<10
45.1
86.9
37.8
549
23.8
<10
30.7
804
285
5240
491
239
677
319
288
557
54.0
21.9
238
20.5
<10
32.5
<10
<10
<10
43.7
18.1
70.6
26.1
<10
39.7
<10
<10
<10
39.7
17.6
95.5
OVCAR-3
NCI/ADR-RES
OVCAR-8
24.0
<10
39.0
22.2
<10
29
<10
<10
<10
<10
<10
<10
39.3
25.8
43.4
533
131
375
29.4
12.9
38.6
1500
789
3490
414
497
2650
414
378
3360
281
1600
403
29.2
23.5
54.7
21.6
<10
34.9
34.6
32.3
63.6
29.1
31.0
43.7
27.8
<10
47.5
28.7
31.4
56.2
Ovarian Cancer
Renal Cancer
A498
RXF 393
SN12C
1720
18.8
35.3
2720
15.9
46.9
1590
<10
<10
4110
<10
<10
1630
208
93.2
28.6
29.4
273
19.1
25.3
72.9
2140
2470
5160
3000
1540
3390
3190
1120
2790
1800
246
656
1720
39.1
58.3
136
15.0
14.0
20.0
27.2
93.3
1260
187
123
2340
33.7
59.8
<10
22.4
88.7
PC-3
DU-145
<10
27.6
26.4
20.9
<10
<10
<10
<10
41.9
38.2
45.2
48.6
37.1
33.8
3720
2080
2830
1830
562
885
275
283
36.1
38.9
<10
25.7
44.8
38.8
41.7
37.7
11.3
26.4
46.3
38.9
Prostate
Cancer
SF-539
SNB-75
SF-295
<10
<10
38.9
15.1
-
19.9
<10
<10
<10
<10
<10
46.3
20.6
27.5
944
268
42.5
46.4
35.5
31.5
16.6
2470
1740
2620
805
528
886
637
534
524
153
30.9
1230
24.0
21.0
54.7
<10
<10
31.3
27.2
21.7
31.6
23.6
36.2
512
<10
<10
39.8
28.5
20.4
29.8
CNS Cancer
HS 578T
BT-549
MCF7
17.3
<10
18.8
<10
<10
26.4
<10
<10
<10
>10000 25.3
80.3
60.1
445
60.1
74.5
19.8
1600
3130
3040
556
996
880
-5.77
524
465
516
-5.86
253
303
385
-6.34
31.7
39.5
35.5
-6.81
20.8
<10
<10
27.2
63.1
59.8
-7.14
26.2
34.7
148
16.7
12.3
24.4
-7.13
26.0
70.8
73.0
-7.12
Breast Cancer
Log10GI50-MIDd
<10
<10
18.2
206
-7.25 -7.17 -7.41 -7.28
-6.79 -6.61
-7.35 -5.54
-7.42
-6.87
a Data obtained from NCI’s in vitro 60 cell 5-dose screen.
b GI50 is the molar concentration of synthetic compound causing 50% growth inhibition of tumor cells.
c Not tested.
d Average activity parameter over all NCI 60-cancer cell lines for the tested compounds.
.
13

2.3.
Molecular docking
The docking studies were realized with GOLD (Figure 4). The protocol has already been described for tubulin
[16] and we chose 2Q8G pdb entry for PDHK1 due to its good resolution [17]. The binding site was defined
as a sphere of 10 Å around the extracted co-crystalized ligand. To validate the docking, this ligand was
redocked in its own binding site and the docked and crystallographic conformations were closely
superimposed.
Reference phenstatin 3b (Figure 4a) and potent antitubulin agents 3g and 4a have been selected for docking
in the tubulin binding site. Compounds 3g and 4a display a binding mode in tubulin that is an inverted image
of phenstatin 3b, with the linker between the two rings pointing toward the inside of the pocket (Figure 4b
and 4c). Compound 3g is nearly superimposable to compound 4a (Figure 4c).
In the PDHK1 binding site, phenstatin 3b has a near unique pose, with the B cycle in the depth of the binding
site where the free phenol forms a hydrogen bond with Ser 75 (Figure 4d). As for the other cycle, a hydrogen
bond with Gln 61 is found consistently, most often with the middle methoxy and less commonly with the
upper one. Compound 3g places differently, with slightly more than two thirds of the poses arranged with
the ester toward the outside of the binding site and the methoxy on the other cycle (A ring) forming a
hydrogen bond with Ser 75. The ester binds to Gln 61 very rarely. It is nonetheless sound to guess at a
hydrogen between them, not seen here due to the orientation of the residue. This may be due to the rigidity
of its side chain as we have not taken its flexibility into account (Figure 4e). Chloroacetyl-thiophenstatin 4a
has a full half of the 30 poses gathered in a single solution placed in reverse when compared with phenstatin
3b but very similar to chloroacetylbenzophenone 3g (Figure 4f). The three-methoxy cycle is positioned in
the depth of the binding site, with one or two of the oxygens forming a hydrogen bond with Ser 75. The
ester carbonyle is sometimes binding to Gln 61. The DCA is very far placed from other inhibitors in another
pocket.
Val171.A
Val171.A
Val171.A
Cys 665.B
Cys 665.B
Cys 665.B
Lys 678.B
Lys 678.B
Lys 678.B
(a)
(b)
(c)
Thr 74
Phe 66
Thr 74
Phe 66
Thr 74
Phe 66
Ser 75
Ser 75
Ser 75
Gln 61
Gln 61
Gln 61
(d)
(e)
(f)
14

Figure 4. Docking of selected molecules in the tubulin binding site (colchicine site) (a) phenstatin 3b; (b)
chloroacetyl-benzophenone 3g; (c) chloroacetyl-thiophenstatin 4a; Docking of selected molecules in the
PDHK1 binding site: (d) phenstatin 3b; (e) chloroacetyl-benzophenone 3g; (f) chloroacetyl-thiophenstatin 4a.
3. Conclusions
The objective of this work was the design, synthesis and pharmacological evaluation of dual-targeted
anticancer agents inhibiting both tubulin polymerization and pyruvate dehydrogenase kinase 1 (PDHK1).
Among the differences between normal and cancer cells are important expressions of PDHK observed in
various cancers including non-small cell lung cancer (NSCLC), breast and colon cancer. Therefore, intensive
efforts are being made to discover effective PDHK inhibitors [4], [5], [18].
The use of PDHK inhibitors has been disappointing in clinical trials so far. Proving the efficacy of dual
tubulin/PDHK1 inhibitors can be crucial, leading to an innovative strategy for the design of new anti-cancer
compounds.
New molecules with improved anticancer activity were thus developed in the current study. Three families
of compounds (mono, di and trichloroacetyl-loaded benzo(thio)phenones) regrouping more than 30
molecules could be obtained (Schemes 1-3). Their biological evaluation confirmed the obtaining of: i)
inhibitors of the tubulin polymerization, classic cytotoxic agents, ii) inhibitors of pyruvate dehydrogenase
kinase 1, rather cytostatic, iii) dual inhibitors tubulin/PDHK1, and iv) highly cytotoxic compounds.
Concerning the specific tubulin polymerization inhibitors discovered in the current study, new SAR have
been highlighted: 1) a new A ring (3,5-dimethyl-4-methoxyphenyl) has been identified and puts an end to
the dogma that the classical 3,4,5-trimethoxyphenyl ring is considered intangible in maintaining antimitotic
activity and consequently, methyl substituents on the A ring can be tolerated in place of methoxy groups in
positions 3 and 5 (compare 3a and 3g, Table 2); 2) the replacement of the methoxy in position 4 of the A
ring by a methyl while conserving the methoxy substituents in positions 3 and 5 slightly decreased
antimitotic activity (compare benzophenones 3a and 3t, Table 1); 3) The carbonyl bridge between the two
rings A and B of parent phenstatin may be successfully replaced by a thiocarbonyl providing unprecedented
thiophenstatin (compare 3a/4a and 3b/4b, Tables 1 and 2); 4) the phenol group of the parent phenstatin
may be effectively replaced with mono-, di- and trichloroacetates (compare antitubulin activities of
benzophenones 3a, 6a, 7a and phenstatin, Table 1); 5) the substitution by trichloroacetate conferring
enhanced antimitotic effect compared to free phenol analogs.
The inhibitors of PDHK1 discovered in this study and which are not active on tubulin polymerization have
been identified as monocloroacetate-loaded benzophenones 3p (Scheme 1), 3v and 3w (Scheme 2) and
also a dichloroacetate-loaded derivative 6d (Scheme 3). The last three molecules have as common
structural feature the same benzophenone scaffold with a 2,4,5-trifluoro-3-methoxyphenyl as A ring. These
PDHK1 inhibitors showed cytostatic effect and did not underwent GI50 calculation on the NCI-60 cancer cell
panel (see Supplementary section for complete one-dose reports). Interestingly, bis-monochloroacetyl (5)
or dichloroacetyl (8) amides resulted in diminished or complete loss of activity against PDHK1, esters being
consequently more favorable than amides for preserving activity against PDHK1 in this study. As expected,
non-chlorinated benzophenone 3x (Scheme 2) bearing an acetyl substituent had no effect on the target
underlining the importance of the halogen.
Monochloroacetate-loaded benzophenone 3g and thiobenzophenone 4a are dual tubulin/PDHK1 anti-
cancer agents. Dichloroacetate-analogue 6b was completely inactive on both targets. In the same way
trichloroacetate 7a was not able to inhibit PDHK1 while conserving a strong antimitotic effect.
Monochloroacetates were the only tolerated to preserve both activities on tubulin and on PDHK1.
The impact of the chloroacetyl-thiophenstatin 4a (dual inhibitor) and thiophenstatin 4b (specific antitubulin
agent) has been evaluated on the cell cycle of DA1-3b cells. Their effect on the cell cycle was different. The
15

thiophenstatin 4b acted as an antitubulin agent and arrested DA1-3b cells in the G2/M phase while the dual
inhibitor 4a mainly increased the sub-G1 phase population and maintained a diminished G2/M effect
compared to thiophenstatin 4b. This may be related to the different mechanisms of action of studied
molecules. It is interesting to note that dichloroacetate, without specified positive counter ion of the tested
molecule, was also reported to arrest bladder cancer cells in G0–G1 phase [19]. Arresting cancer cells in
subG1 phase is interesting for anticancer drugs, but we do not know if the cells in the subG1 are not in G0
and therefore in quiescence. Quiescence is also encouraging for stopping tumor growth. Indeed, an
inhibition of the production of cyclin D1 may be caused by the dual inhibitor 4a since a strong expression
of cyclin D1 is necessary to pass the control point in phase G1 and start the phase S, but this is an assumption
which deserves further biological efforts in due course.
Excellent cell growth inhibition on the NCI-60 cancer cell lines panel has been registered for our antimitotic
agents and dual tubulin/PDHK1 inhibitors. The most sensitive cells to these molecules proved to be
melanoma (in particular MDA-MB-435 and M14), colon cancer (KM12 and HCT-15) and leukemia (HL-60(TB)
and K-562) cell lines. It is also interesting to note that dual tubulin/PDHK1 inhibitor 4a exhibited improved
cancer cell growth inhibition compared to analogue molecule 3a showing only antitubulin efficacy on non-
small cell lung cancer (A549/ATCC, NCI-H522 and NCI-H322M) and renal A498 cancer cells. This may be
explained by the significantly up-regulated expression of PDHK1 in NSCLC cells including A549 cells [20].
The same tendency has been observed for benzophenones 3g and 3h. The monochloroacetylated
benzophenone 3g, dual inhibitor tubulin/PDHK1, showed improved anticancer potential on A549/ATCC
NSCLC cells and on A498 renal cancer cells compared to benzophenone 3h bearing only a free phenol
substituent on the B ring.
To the best of our knowledge this is the first time when a potential codrug acting on tubulin polymerization
and on PDHK1 is envisaged. This duality of actions offers new perspectives and supports the importance of
such inhibitors in the design of new anticancer drugs.
4. Experimental section
4.1. Materials and methods
Starting materials are commercially available and were used without further purification (suppliers:
Carlo Erba Reagents S.A.S., Tokyo Chemical Industry Co. Ltd. and Acros Organics). Melting points
were measured on a MPA 100 OptiMelt® apparatus and are uncorrected. Nuclear Resonance
Magnetic (NMR) were acquired at 400 MHz for 1H NMR, at 376 MHz for 19F NMR and at 100 MHz
for 13C NMR, on a Varian 400-MR spectrometer with tetramethylsilane (TMS) as internal standard,
at 25°C. Chemical shifts (δ) are expressed in ppm relative to TMS. Splitting patterns are designed:
s, singlet; d, doublet; dd, doublet of doublet; t, triplet; m, multiplet; quint, quintuplet; br s, broaden
singlet; br t, broaden triplet. Coupling constants (J) are reported in Hertz (Hz). Thin layer
chromatography (TLC) was realized on Macherey Nagel silica gel plates with fluorescent indicator
and were visualized under a UV-lamp at 254 nm and 365 nm. Column chromatography was
performed with a CombiFlash Rf Companion (Teledyne-Isco System) using RediSep packed columns.
IR spectra were recorded on a Varian 640-IR FT-IR Spectrometer. Elemental analyses (C, H, N) of
new compounds were determined on a Thermo Electron apparatus by “Pôle Chimie Moléculaire-
Welience”, Faculté des Sciences Mirande, Dijon, France.
4.1.1. General procedure A for Friedel–Crafts reactions in the presence of Eaton’s reagent
Eaton’s reagent was prepared from phosphorus pentoxide (P2O5) and methanesulfonic acid (CH3SO3H)
(weight ratio P2O5/CH3SO3H 1:10). The mixture was heated at 40 °C under nitrogen atmosphere until
complete homogeneity. Benzoic acid (1.15–1.5 equiv) and aromatic derivative (1.0 equiv) were then added
to Eaton’s reagent. The mixture was heated at 60°C under inert atmosphere for 3–24 h. After cooling to
room temperature, the reaction medium was diluted with dichloromethane and carefully poured into a
separatory funnel containing sodium bicarbonate aqueous solution (50% NaHCO3) (neutralization to pH 7).
The aqueous solution was extracted with dichloromethane, and the combined organic layers were dried
(MgSO4). Solvent was removed under reduced pressure to produce a brownish oil. The crude product was
16

purified by Flash chromatography on RediSep packed columns (eluent n-heptane/ethyl acetate) to provide
pure benzophenones 3g, 3i, 3k and 3o.
Benzophenones 3a, 3b, 3c, 3e, 3m, 3p, 3q, 3r, 3t and 3x have been previously reported and re-synthesized
accordingly [7], [8], [11], [12], [13].
4.1.1.1. 2-Methoxy-5-(4-methoxy-3,5-dimethylbenzoyl)phenyl chloroacetate (3g). General procedure A was
followed by using 4-methoxy-3,5-dimethylbenzoic acid 16 (1.45 g, 8.05 mmol), 2-methoxyphenyl
chloroacetate 20 (1.08 g, 5.38 mmol) and Eaton’s reagent (0.53 g P2O5 in 3.6 mL CH3SO3H). The crude
obtained after work-up was purified on silica gel column (eluent: gradient EtOAc/n-heptane 3/7 → 100%
EtOAc) to provide pure benzophenone 3g as a beige solid; 80% yield (1.56 g); mp (EtOAc) 76-78 °C; 1H NMR
(CDCl3, 400MHz) δ (ppm) 2.33 (s, 6H, 2ArCH3), 3.78 (s, 3H, OCH3), 3.92 (s, 3H, OCH3), 4.35 (s, 2H, OCOCH2Cl),
7.04 (d, J = 8.8 Hz, 1H, ArH), 7.45 (s, 2H, ArH), 7.60 (d, J = 2.1 Hz, 1H, ArH), 7.75 (dd, J = 8.8, 2.1 Hz, 1H, ArH).
13C NMR (CDCl3, 100 MHz) δ 16.1 (2CH3), 40.5 (CH2), 56.1 (CH3), 59.6 (CH3), 111.4 (CH), 111.6 (CH), 124.3
(2CH), 124.4 (2C), 133.0 (CH), 133.2 (C), 138.8 (C), 139.5 (C), 154.2 (C), 160.5 (C), 165.2 (C), 193.8 (C). IR ν
(cm-1): 1736, 1630, 1593, 1506, 1441, 1265, 1223, 1141, 1116, 1003, 882, 752. Elem. Analysis calcd. for
C19H19ClO5: C, 62.90; H, 5.28. Found: C, 62.81; H, 5.13%.
4.1.1.2. 2,6-Dimethoxy-3-(4-methoxy-3,5-dimethylbenzoyl)phenyl chloroacetate (3i). General procedure A
was followed by using 4-methoxy-3,5-dimethylbenzoic acid 16 (1.00 g, 5.55 mmol), 2,6-dimethoxyphenyl
chloroacetate 21 (0.85 g, 3.7 mmol) and Eaton’s reagent (0.63 g P2O5 in 2.46 mL CH3SO3H). The crude
obtained after work-up was purified on silica gel column (eluent: gradient EtOAc/n-heptane 3/7 → 100%
1
EtOAc) to provide pure benzophenone 3i as a white solid; 87% yield (1.26 g); mp (EtOAc) 104-107 °C; H
NMR (CDCl3, 400MHz) δ (ppm) 2.33 (s, 6H, 2ArCH3), 3.82 (s, 3H, OCH3), 3.95 (s, 3H, OCH3), 4.36 (s, 2H,
OCOCH2Cl), 7.06 (d, J = 8.6 Hz, 1H, ArH), 7.42 (s, 2H, ArH), 7.62 (d, J = 8.6 Hz, 1H, ArH). 13C NMR (CDCl3, 100
MHz) δ 16.2 (2CH3), 40.4 (CH2), 56.3 (CH3), 60.0 (CH3), 62.5 (CH3), 110.9 (CH), 111.1 (CH), 124.2 (2CH), 124.5
(2C), 132.9 (CH), 133.1 (C), 138.9 (C), 139.6 (C), 154.3 (C), 160.6 (C), 165.3 (C), 193.7 (C). Elem. Analysis calcd.
for C20H21ClO6: C, 61.15; H, 5.39. Found: C, 61.40; H, 5.62%.
4.1.1.3. N-[2,6-Dimethoxy-3-(4-methoxy-3,5-dimethylbenzoyl)phenyl]acetamide (3k). General procedure A was
followed by using 4-methoxy-3,5-dimethylbenzoic acid 16 (1.62 g, 8.99 mmol), N-(2,6-
dimethoxyphenyl)acetamide 22 (1.17 g, 5.99 mmol) and Eaton’s reagent (0.59 g P2O5 in 3.98 mL CH3SO3H).
The crude obtained after work-up was purified on silica gel column (eluent: EtOAc/n-heptane 6/4) to provide
1
pure benzophenone 3k as a white solid in 66% yield (1.41 g); mp (EtOAc/n-heptane) 188-190 °C; H NMR
(CDCl3, 400MHz) δ (ppm) 2.18 (s large, 3H, NHCOCH3), 2.31 (s, 6H, 2ArCH3), 3.65 (s, 3H, OCH3), 3.77 (s, 3H, OCH3),
3.91 (s, 3H, OCH3), 6.68 (s large, 1H, ArNH), 6.75 (d, J = 8.9 Hz, 1H, ArH), 7.30 (d, J = 8.9 Hz, 1H, ArH), 7.52 (s, 2H, ArH).
13C NMR (CDCl3, 100 MHz) δ 16.2 (2CH3), 23.3 (CH3), 56.1 (CH3), 60.0 (CH3), 62.1 (CH3), 106.1 (C), 119.1 (C), 125.7
(C), 129.4 (CH), 131.0 (CH), 131.1 (2CH), 133.3 (2C), 155.5 (C), 157.4 (C), 161.3 (C), 173.9 (C), 194.5 (C). Elem. Analysis
calcd. for C20H23O5N: C, 67.21; H, 6.49; N, 3.92. Found: C, 67.01; H, 6.70; N, 3.92%.
4.1.1.4. (2,3-Dihydroxy-4-methoxyphenyl)(2,4,5-trifluoro-3-methoxyphenyl)methanone (3o). General
procedure A was followed by using 2,4,5-trifluoro-3-methoxybenzoic acid 17 (2.5 g, 12.1 mmol), 2,6-
dimethoxyphenyl chloroacetate 21 (1.87 g, 8.09 mmol) and Eaton’s reagent (0.91 g P2O5 in 6.14 mL
CH3SO3H). The crude obtained after work-up was purified on silica gel column (eluent: EtOAc/n-heptane 3/7)
1
to provide pure benzophenone 3o as a yellow solid; 71% yield (1.89 g); mp (EtOAc) 120-123 °C; H NMR
(CDCl3, 400MHz) δ (ppm 3.98 (s, 3H, OCH3), 4.09 (s, 3H, OCH3), 5.54 (s, 1H, ArOH), 6.51 (d, J = 9.2 Hz, 1H,
ArH), 6.97 (sm, 1H, ArH), 6.98 (d, J = 9.2 Hz, 1H, ArH), 11.89 (s, 1H, ArOH). 19F NMR (CDCl3, 376 MHz) δ (ppm)
-146.37 (ddd, J = 20.5, 12.4, 9.6 Hz, 1F, ArF), -138.92 (dddd, J = 22.4, 20.5, 13.5, 9.5 Hz, 1F, ArF), -132.42
(sm, 1F, ArF). 13C NMR (CDCl3, 100 MHz) δ 56.4 (CH3), 62.2 (t, J = 3.1 Hz, CH3), 103.3 (CH), 109.9 (dd, J = 20.2,
3.1 Hz, CH), 114.4 (C), 121.9 (sm, C), 125.5 (d, J = 2.4 Hz, CH), 133.5 (C), 138.3 (m, C), 145.9 (ddd, J = 255.0,
15.0, 5.3 Hz, C), 147.4 (ddd, J = 248.5, 11.7, 3.1 Hz, C), 148.5 (dt, J = 249.6, 3.3 Hz, C), 150.8 (C), 153.1 (C),
17

194.3 (C). IR ν (cm-1): 3467, 1633, 1597, 1508, 1463, 1429, 1330, 1278, 1095, 1021, 777, 675, 499. Elem.
Analysis calcd. for C15H11O5F3: C, 54.89; H, 3.38. Found: C, 54.77; H, 3.17%.
4.1.2. General procedure B: synthesis of phenols from monochloroacetic esters
Monochloroacetic ester 3g and 3i (1.0 equiv) and sodium acetate (AcONa·3H2O; 4.5 equiv) were dissolved
in MeOH. The solution was heated to reflux for 5 h. After cooling to RT, the mixture was concentrated under
reduced pressure and the residue was taken into distilled water. The resulting precipitate was filtered,
washed with water several times to remove remaining sodium acetate, and recrystallized from absolute
EtOH or purified by flash chromatography to obtain pure benzophenones bearing free phenols 3h and 3j.
4.1.2.1. (3-Hydroxy-4-methoxyphenyl)(4-methoxy-3,5-dimethylphenyl)methanone (3h). General procedure
B was followed by using monochloroacetate 3g (0.86 g, 2.37 mmol) and sodium acetate (AcONa·3H2O; 1.45
g, 10.7 mmol) in MeOH (25 mL). After evaporation of the methanol in vacuo, the residue was poured in
distilled water. The residue was purified by column chromatography (EtOAc/n-heptane, 15:85) and
recrystallized from EtOH to obtain benzophenone 3h as a white solid; 92% yield (0.62 g); mp (EtOAc) 118-
119 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 2.33 (s, 6H, 2ArCH3), 3.78 (s, 3H, OCH3), 3.98 (s, 3H, OCH3), 5.67
(s, 1H, ArOH), 6.92 (d, J = 8.4 Hz, 1H, ArH), 7.38 (dd, J = 8.3, 2.2 Hz, 1H, ArH), 7.41 (d, J = 2.1 Hz, 1H, ArH),
13
7.45 (s, 2H, ArH). C NMR (CDCl3, 100 MHz) δ 16.2 (2CH3), 56.1 (CH3), 59.7 (CH3), 109.7 (CH), 116.3 (CH),
123.6 (CH), 130.8 (C), 131.2 (2C), 131.4 (2CH), 139.7 (C), 145.2 (C), 150.0 (C), 160.6 (C), 195.2 (C). IR ν (cm-
1): 3226, 1631, 1603, 1507, 1442, 1313, 1224, 1174, 1141, 1117, 1005, 753. Elem. Analysis calcd. for
C17H18O4: C, 71.31; H, 6.34. Found: C, 71.61; H, 5.93%.
4.1.2.2. (3-Hydroxy-2,4-dimethoxyphenyl)(4-methoxy-3,5-dimethylphenyl)methanone (3j). General
procedure B was followed by using monochloroacetate 3i (0.98 g, 2.5 mmol) and sodium acetate
(AcONa·3H2O; 1.45 g, 35.7 mmol) in MeOH (35 mL). After evaporation of the methanol in vacuo, the residue
was poured in distilled water. The resulting solid was collected by filtration and recrystallized from absolute
EtOH to obtain pure benzophenone 3j as a sparkling white solid; 98% yield (0.77 g); mp (EtOAc) 110-112 °C;
1H NMR (CDCl3, 400MHz) δ (ppm) 2.30 (s, 6H, 2ArCH3), 3.76 (s, 3H, OCH3), 3.77 (s, 3H, OCH3), 3.95 (s, 3H,
OCH3), 5.79 (s large, 1H, ArOH), 6.70 (d, J = 8.6 Hz, 1H, ArH), 6.90 (d, J = 8.6 Hz, 1H, ArH), 7.50 (s, 2H, ArH).
13C NMR (CDCl3, 100 MHz) δ 16.2 (2CH3), 56.3 (CH3), 59.7 (CH3), 61.8 (CH3), 105.9 (CH), 120.8 (CH), 126.1 (C),
130.9 (2C), 131.0 (2CH), 133.6 (C), 138.3 (C), 145.8 (C), 149.7 (C), 161.2 (C), 194.9 (C). Elem. Analysis calcd.
for C18H20O5: C, 68.34; H, 6.37. Found: C, 67.96; H, 6.09%.
4.1.3. (3-Amino-2,4-dimethoxyphenyl)(4-methoxy-3,5-dimethylphenyl)methanone (3l). A mixture of
acetamide 3k (1.52 g, 4.25 mmol) and a 10% hydrochloric acid solution (30 mL) in methanol (20 mL) was
stirred at reflux for 24 h. The final suspension was neutralized to pH 7 by slow addition of an ammoniac
solution. The aqueous solution was extracted with dichloromethane. The organic phase was dried (MgSO4),
filtered and concentrated in vacuo. The residue was purified by liquid chromatography on silica column
1
(dichloromethane) to afford pure aniline 3l as a yellow cotton (1.14 g, 85% yield); mp (CH2Cl2) 139 °C; H
NMR (CDCl3, 400MHz) δ (ppm) 2.29 (s, 6H, 2ArCH3), 3.69 (s, 3H, OCH3), 3.77 (s, 3H, OCH3), 3.92 (s, 3H, OCH3), 3.99
(s large, 2H, ArNH2), 6.63 (d, J = 8.5 Hz, 1H, ArH), 6.75 (d, J = 8.5 Hz, 1H, ArH), 7.52 (s, 2H, ArH); 13C NMR (CDCl3, 100
MHz) δ 16.2 (2CH3), 55.8 (CH3), 59.6 (CH3), 61.2 (CH3), 105.2 (C), 119.0 (CH), 125.7 (C), 129.6 (C), 130.8 (2C), 131.1
(2CH), 133.7 (C), 145.7 (C), 149.9 (C), 161.1 (C), 195.4 (C). Elem. Analysis calcd. for C18H21O4N: C, 68.55; H, 6. 71;
N, 4.44. Found: C, 68.46; H, 6.93; N, 4.34%.
4.1.4. 2-Methoxy-5-[(3,4,5-trimethoxyphenyl)carbonothioyl]phenyl chloroacetate (4a). Phosphorous
pentasulfide (0.92 g, 4.05 mmol) and hexamethyledisiloxane (0.92 mL, 8.1 mmol) were added to a solution
of chloroacetic-phenstatine 3a (1.6 g, 4.05 mmol) in 25 mL toluene and the mixture was stirred at 110 °C
under inert atmosphere for 24 hours and then concentrated under vacuum. The residue was purified by
silica gel column chromatography (n-heptane:ethyl acetate from 100:0 to 0:100) to give 4a as a blue-green
solid (1.05 g, 63% yield); mp (EtOAc) 90-91 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.89 (s, 9H, 3OCH3), 3.94
(s, 3H, OCH3), 4.37 (s, 2H, OCOCH2Cl), 7.05 (s, 2H, 2ArH), 7.09 (d, J = 7.6 Hz, 1H, ArH), 7.60 (d, J = 1.5 Hz, 1H,
ArH), 7.86 (dd, J = 7.6, 1.5 Hz, 1H, ArH). 13C NMR (CDCl3, 100 MHz) δ 40.5 (CH2), 56.2 (CH3), 56.3 (2CH3), 60.9
18

(CH3), 107.6 (2CH), 111.4 (CH), 124.7 (CH), 130.1 (CH), 138.5 (C), 140.2 (C), 141.9 (C), 142.5 (C), 152.4 (2C),
154.5 (C), 165.4 (C), 232.6 (C). IR ν (cm-1): 1737, 1601, 1577, 1505, 1488, 1407, 1329, 1287, 1183, 1123, 989,
829, 791. Elem. Analysis calcd. for C19H19O6SCl: C, 55.54; H, 4.66; S, 7.80. Found: C, 55.59; H, 4.68; S, 7.54%.
4.1.5.
(3-Hydroxy-4-methoxyphenyl)(3,4,5-trimethoxyphenyl)methanethione
(4b).
Phosphorous
pentasulfide (1.0 g, 4.4 mmol) and hexamethyledisiloxane (1.1 mL, 8.8 mmol) were added to a solution of
phenstatine 3b (1.39 g, 4.4 mmol) in 25 mL toluene and the mixture was stirred at 110°C under inert
atmosphere for 24 hours and then concentrated under vacuum. The residue was purified by silica gel
column chromatography (n-heptane:ethyl acetate from 100:0 to 100:0) to give 4b as a blue solid (769 mg,
43% yield). 1H NMR (CDCl3, 400 MHz) δ (ppm): 3.86 (s, 6H, 2OCH3), 3.94 (s, 3H, OCH3), 3.99 (s, 3H, OCH3),
5.62 (br s, 1H, OH), 6.84 (d, J = 8.7 Hz, 1H, ArH), 6.96 (s, 2H, ArH), 7.34 (dd, J = 8.4, 2.1 Hz, 1H, ArH), 7.44 (d,
13
J = 2.1 Hz, 1H, ArH). C NMR (CDCl3, 100 MHz) δ (ppm) 56.1 (CH3), 56.3 (2CH3), 60.9 (CH3), 107.7 (2CH),
109.3 (CH), 116.6 (CH), 123.5 (CH), 141.2 (C), 141.7 (C), 143.0 (C), 145.0 (C), 150.5 (C), 152.3 (2C), 234.2 (C).
IR ν (cm-1): 3448, 1572, 1499, 1451, 1320, 1276, 1123, 1026, 983, 849. Elem. Analysis calcd. for C17H18O5S: C,
61.06; H, 5.43; S, 9.59. Found: C, 61.31; H, 5.49; S, 9.28%.
4.1.6. General procedure C for the synthesis of mono-, di- and trichloroacetic esters and amides from phenols
and anilines
A mixture of phenol (3b, 3h, 3j or 3o) or aniline 3l (1 equiv) and monochloroacetic acid chloride,
dichloroacetic acid chloride or trichloroacetic acid chloride (2-6 equiv) was placed under nitrogen
atmosphere and heated at reflux for 1-6 h. Once the reaction started, a vigorous release of gaseous
hydrochloric acid was observed. After reaction completion, the mixture was cooled to room temperature,
treated with a saturated solution NaHCO3 and extracted several times with dichloromethane. Combined
organic phases were dried over magnesium sulfate and then concentrated. The oily resulting residues were
crystallized from absolute ethanol or purified by column chromatography to afford pure protected phenols
3v, 3w, 6a, 6b, 6c, 6d, 7a or 7b or amides 5 or 8.
4.1.6.1. 3-Methoxy-6-(2,4,5-trifluoro-3-methoxybenzoyl)-1,2-phenylene bis(2-chloroacetate) (3v). General
procedure C was followed by using benzophenone 3o (0.1 g, 3.08 mmol) and monochloroacetic acid
chloride (1.47 mL, 18.48 mmol). The crude obtained after final work-up was purified by flash
chromatography (eluent: EtOAc/n-heptane 6/4) to provide pure bischloroacetate 3v as a white solid in 41%
yield (0.06 g); mp (EtOAc/n-heptane) 103-105 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.94 (s, 3H, OCH3), 4.06
(s, 3H, OCH3), 4.26 (s, 2H, OCOCH2Cl), 4.34 (s, 2H, OCOCH2Cl), 6.93 (d, J = 8.9 Hz, 1H, ArH), 7.10 (sym m, 1H,
ArH), 7.49 (dd, J = 9.0, 1.9 Hz, 1H, ArH). 19F NMR (CDCl3, 376 MHz) – 144.40 (sym m, 1F, ArF), - 138.71 (sym
m, 1F, ArF), -130.69 (sym m, 1F, ArF). 13C NMR (CDCl3, 100 MHz) δ 40.1 (CH2), 40.3 (CH2), 56.7 (CH3), 62.2
(CH3), 109.2 (CH), 110.8 (d, CH), 123.3 (C), 130.7 (CH), 132.4 (C), 142.7 (C), 156.2 (C), 164.1 (C), 164.7 (C),
187.0 (C). IR ν (cm-1): 1795, 1661, 1608, 1505, 1452, 1407, 1240, 1121, 1095, 1025, 929, 820, 782, 747, 631.
Elem. Analysis calcd. for C19H13Cl2F3O7: C, 47.42; H, 2.72. Found: C, 47.78; H, 2.93%.
4.1.6.2. 2-Hydroxy-6-methoxy-3-(2,4,5-trifluoro-3-methoxybenzoyl)phenyl 2-chloroacetate (3w). By-
product from the synthesis of dichloroacetate 3v. White solid; 37% yield (0.046 g); mp (EtOAc/n-heptane)
108-110 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.92 (s, 3H, OCH3), 4.10 (s, 3H, OCH3), 4.43 (s, 2H, OCOCH2Cl),
6.54 (d, J = 9.0 Hz, 1H, ArH), 6.98 (sym m, 1H, ArH), 7.32 (dd, J = 9.0, 2.8 Hz, 1H, ArH), 12.10 (s, 1H, ArOH).
13C NMR (CDCl3, 100 MHz) δ 40.4 (CH2), 56.5 (CH3), 61.6 (CH3), 103.3 (CH), 100.0 (CH), 109.8 (dd, C), 114.6
(C), 122.1 (C), 125.5 (C), 127.0 (C), 132.4 (CH), 144.8 (dd, C), 147.3 (d, C), 156.3 (C), 158.2 (C), 164.9 (C),
194.1 (C). IR ν (cm-1): 3062, 1785, 1639, 1505, 1471, 1411, 1370, 1354, 1322, 1281, 1254, 1127, 1011, 945,
926, 811, 782.
4.1.6.3.
2-Chloro-N-(2-chloro-acetyl)-N-[2,6-dimethoxy-3-(4-methoxy-3,5-dimethyl-benzoyl)-phenyl]-
acetamide (5). General procedure C was followed by using benzophenone 3l (0.3 g, 0.95 mmol) and
monochloroacetic acid chloride (0.45 mL, 5.7 mmol). The oily resulting residue was crystallized from
19

absolute ethanol to provide bis-monochloroacetamide 5 as a white solid; 98% yield (0.44 g); mp (EtOH) 152-
155 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 2.32 (s, 6H, 2CH3), 3.57 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 3.90 (s,
3H, OCH3), 4.47 (q, J = 23.1, 16.1 Hz, 4H, 2COCHCl2), 6.80 (d, J = 9.0 Hz, 1H, ArH), 7.49 (s, 2H, ArH), 7.50 (d,
J = 9.0 Hz, 1H, ArH). 13C NMR (CDCl3, 100 MHz) δ 16.3 (2CH3), 45.2 (2CH2), 56.6 (CH3), 59.7 (CH3), 62.4 (CH3),
106.2 (CH), 118.0 (C), 125.0 (C), 131.0 (2CH), 131.4 (2C), 132.8 (CH), 133.2 (C), 156.4 (C), 157.6 (C), 161.7
(C), 168.6 (2C), 193.7 (C). IR ν (cm-1): 1713, 1653, 1594, 1462, 1415, 1317, 1292, 1206, 1125, 1097, 1007,
793. Elem. Analysis calcd. for C22H23Cl2O6N: C, 56.42; H, 4.95; N, 2.99. Found: C, 56.21; H, 5.07; N, 3.14%.
4.1.6.4. 2-Methoxy-5-(3,4,5-trimethoxybenzoyl)phenyl 2,2-dichloroacetate (6a). General procedure C was
followed by using phenstatine 3b (0.5 g, 1.57 mmol) and dichloroacetic acid chloride (0.29 mL, 3.14 mmol).
The oily resulting residue was crystallized from absolute ethanol to provide dichloroacetate 6a as a white
solid; 95% yield (0.64 g); mp (EtOH) 104-107 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.89 (s, 6H, 2OCH3), 3.94
(s, 3H, OCH3), 3.95 (s, 3H, OCH3), 6.21 (s, 1H, OCOCHCl2), 7.03 (s, 2H, ArH), 7.08 (d, J = 8.6 Hz, 1H, ArH), 7.66
(d, J = 1.96 Hz, 1H, ArH), 7.81 (dd, J = 8.6, 1.96 Hz, 1H, ArH). 13C NMR (CDCl3, 100 MHz) δ 56.2 (CH3), 56.3
(2CH3), 61.0 (CH3), 63.8 (CH), 107.4 (2CH), 111.8 (CH), 124.3 (CH), 130.4 (C), 130.6 (CH), 132.5 (C), 138.5 (C),
141.9 (C), 152.9 (2C), 154.3 (C), 162.5 (C), 193.4 (C). IR ν (cm-1): 1780, 1641, 1607, 1580, 1502, 1425, 1330,
1269, 1228, 1125, 997, 834, 760. Elem. Analysis calcd. for C19H18Cl2O7: C, 53.16; H, 4.23. Found: C, 53.31; H,
4.27%.
4.1.6.5. 2-Methoxy-5-(4-methoxy-3,5-dimethylbenzoyl)phenyl 2,2-dichloroacetate (6b). General procedure
C was followed by using benzophenone 3h (0.5 g, 1.75 mmol) and dichloroacetic acid chloride (0.34 mL, 3.5
mmol). The oily resulting residue was crystallized from absolute ethanol to provide dichloroacetate 6b as a
beige solid; 77% yield (0.53 g); mp (EtOAc/n-heptane) 113-116 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 2.34
(s, 6H, 2CH3), 3.79 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 6.21 (s, 1H, OCOCHCl2), 7.06 (d, J = 8.6 Hz, 1H, ArH), 7.45
(s, 2H, ArH), 7.65 (d, J = 2.0 Hz, 1H, ArH), 7.77 (dd, J = 8.6, 2.0 Hz, 1H, ArH). 13C NMR (CDCl3, 100 MHz) δ 16.2
(2CH3), 56.3 (CH3), 59.7 (CH3), 63.8 (CH), 111.6 (CH), 124.1 (CH), 130.6 (C), 130.8 (CH), 130.9 (2CH), 131.1
(2C), 133.0 (C), 138.6 (C), 154.2 (C), 160.7 (C), 162.4 (C), 194.1 (C). IR ν (cm-1): 1743, 1603, 1574, 1506, 1454,
1438, 1350, 1314, 1264, 1217, 1194, 1169, 1135, 1002, 890, 804, 774, 754, 624. Elem. Analysis calcd. for
C19H18Cl2O5: C, 57.45; H, 4.57. Found: C, 57.80; H, 4.92%.
4.1.6.6. 2,6-Dimethoxy-3-(4-methoxy-3,5-dimethylbenzoyl)phenyl 2,2-dichloroacetate (6c). General
procedure C was followed by using benzophenone 3j (0.1 g, 0.32 mmol) and dichloroacetic acid chloride
(0.06 mL, 0.64 mmol). The crude obtained after final work-up was purified by flash chromatography (eluent:
EtOAc/n-heptane 6/4) to provide pure dichloroacetate 6c as a beige oil in 88% yield (0.12 g); 1H NMR (CDCl3,
400MHz) δ (ppm) 2.31 (s, 6H, 2ArCH3), 3.76 (s, 3H, OCH3), 3.77 (s, 3H, OCH3), 3.95 (s, 3H, OCH3), 5.96 (s, 1H,
OCOCHCl2), 6.80 (d, J = 8.6 Hz, 1H, ArH), 7.32 (d, J = 8.6 Hz, 1H, ArH), 7.50 (s, 2H, ArH). 13C NMR (CDCl3, 100
MHz) δ 16.2 (2CH3), 56.3 (CH3), 59.7 (CH3), 61.8 (CH3), 64.2 (CH), 105.8 (CH), 120.9 (CH), 125.9 (C), 130.9
(2C), 131.1 (2CH), 133.4 (C), 138.3 (C), 145.8 (C), 149.8 (C), 154.8 (C), 161.2 (C), 195.2 (C). Elem. Analysis
calcd. for C20H20Cl2O6: C, 56.22; H, 4.72. Found: C, 56.45; H, 5.00%.
4.1.6.7. 3-Methoxy-6-(2,4,5-trifluoro-3-methoxybenzoyl)1,2-phenylene bis(2,2-dichloroacetate) (6d).
General procedure C was followed by using benzophenone 3o (0.5 g, 1.52 mmol) and monochloroacetic
acid chloride (0.88 mL, 9.14 mmol). The crude obtained after final work-up was purified by flash
chromatography (eluent: EtOAc/n-heptane 6/4) to provide pure bisdichloroacetate 6d as a beige solid in 63%
yield (0.53 g); mp (EtOAc/n-heptane) 107-109 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.97 (s, 3H, OCH3), 4.05
(t, J = 1.1 Hz, 3H, OCH3), 6.15 (s, 1H, OCOCHCl2), 6.20 (s, 1H, OCOCHCl2), 6.99 (d, J = 8.6 Hz, 1H, ArH), 7.12
(sym m, J = 8.6, 6.2 Hz, 1H, ArH), 7.54 (dd, J = 9.0, 2.0 Hz, 1H, ArH). 19F NMR (CDCl3, 376 MHz) – 143.84 (sym
m, 1F, ArF), - 138.63 (sym m, 1F, ArF), -130.22 (sym m, 1F, ArF). 13C NMR (CDCl3, 100 MHz) δ 56.8 (CH3), 62.2
(CH3), 63.3 (CH), 63.5 (CH), 109.8 (CH), 110.9 (dd, J = 19.8, 2.8 Hz, CH), 122.1 (sym m, C), 123.5 (C), 130.9
(CH), 131.8 (C), 142.1 (C), 146.0 (sym m, C), 148.5 (sym m, C), 148.8 (sym m, C), 151.4 (sym m, C), 156.0 (C),
20

161.1 (C), 161.5 (C), 186.6 (C). IR ν (cm-1): 1799, 1645, 1611, 1506, 1466, 1354, 1292, 1255, 1118, 1090,
1022, 935, 829, 771. Elem. Analysis calcd. for C19H11Cl4F3O7: C, 41.49; H, 2.02. Found: C, 41.81; H, 2.29%.
4.1.6.8. 2-Methoxy-5-(3,4,5-trimethoxybenzoyl)phenyl 2,2,2-trichloroacetate (7a). General procedure C was
followed by using phenstatine 3b (0.5 g, 1.57 mmol) and trichloroacetic acid chloride (0.29 mL, 3.14 mmol).
The oily resulting residue was crystallized from absolute ethanol to provide trichloroacetate 7a as a grey
solid; 84% yield (0.61 g); mp (Et2O) 139-142 °C; 1H NMR (CDCl3, 400MHz) δ (ppm) 3.89 (s, 6H, 2OCH3), 3.95
(s, 3H, OCH3), 3.97 (s, 3H, OCH3), 7.03 (s, 2H, ArH), 7.10 (d, J = 8.6 Hz, 1H, ArH), 7.71 (d, J = 1.9 Hz, 1H, ArH),
7.82 (dd, J = 8.6, 1.9 Hz, 1H, ArH). 13C NMR (CDCl3, 100 MHz) δ 56.3 (2CH3), 56.4 (CH3), 61.0 (CH3), 89.4 (C),
107.4 (2CH), 111.9 (CH), 124.1 (CH), 130.5 (C), 130.9 (CH), 132.5 (C), 138.9 (C), 141.9 (C), 153.0 (2C), 154.3
(C), 160.1 (C), 193.4 (C). IR ν (cm-1): 1782, 1650, 1607, 1583, 1503, 1414, 1332, 1276, 1130, 1178, 1016, 970,
843, 763, 679. Elem. Analysis calcd. for C19H17Cl3O7: C, 49.22; H, 3.70. Found: C, 49.63; H, 4.01%.
4.1.6.9. 2-Methoxy-5-(4-methoxy-3,5-dimethylbenzoyl)phenyl 2,2,2-trichloroacetate (7b). General
procedure C was followed by using benzophenone 3h (0.5 g, 1.75 mmol) and trichloroacetic acid chloride
(0.39 mL, 3.50 mmol). The crude obtained after final work-up was purified by flash chromatography (eluent:
1
EtOAc/n-heptane 6/4) to provide trichloroacetate 7b as a beige oil; 98% yield (0.74 g); H NMR (CDCl3,
400MHz) δ (ppm) 2.32 (s, 6H, 2CH3), 3.80 (s, 3H, OCH3), 3.91 (s, 3H, OCH3), 7.17 (d, J = 8.6 Hz, 1H, ArH), 7.46
(s, 2H, ArH), 7.72 (d, J = 2.0 Hz, 1H, ArH), 7.80 (dd, J = 8.6, 2.0 Hz, 1H, ArH). Elem. Analysis calcd. for
C19H17Cl3O5: C, 52.86; H, 3.97. Found: C, 53.09; H, 4.27%.
4.1.6.10.
2,2-Dichloro-N-(2,2-dichloro-acetyl)-N-[2,6-dimethoxy-3-(4-methoxy-3,5-dimethyl-benzoyl)-
phenyl]-acetamide (8). General procedure C was followed by using benzophenone 3l (0.3 g, 0.95 mmol) and
dichloroacetic acid chloride (0.55 mL, 5.7 mmol). The oily resulting residue was crystallized from absolute
ethanol to provide benzophenone 8 as a white solid; 98% yield (0.50 g); mp (EtOAc/n-heptane) 129-131 °C;
1H NMR (CDCl3, 400MHz) δ (ppm) 2.32 (s, 6H, 2CH3), 3.57 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 3.91 (s, 3H,
OCH3), 6.62 (s, 2H, 2COCHCl2), 6.83 (d, J = 9.0 Hz, 1H, ArH), 7.50 (s, 2H, ArH), 7.55 (d, J = 9.0 Hz, 1H, ArH).
13C NMR (CDCl3, 100 MHz) δ 16.3 (2CH3), 56.6 (CH3), 59.7 (CH3), 62.4 (CH3), 65.8 (CH), 106.4 (CH), 117.1 (C),
125.0 (C), 131.0 (2CH), 131.5 (2C), 132.6 (C), 133.7 (CH), 156.0 (C), 157.5 (C), 161.9 (C), 166.0 (C), 193.5 (C).
Elem. Analysis calcd. for C22H21Cl4O6N: C, 49.19; H, 3.94. Found: C, 49.50; H, 4.15%.
4.1.7. 2-Methoxy-5-(3,4,5-trimethoxybenzoyl)phenyl acetate (3x). A mixture of phenstatin 3b and acetyl
chloride was heated at reflux during 3h. After cooling to RT, the reaction medium was diluted with
dichloromethane and carefully poured into a separatory funnel containing aqueous sodium bicarbonate
solution (10% NaHCO3). The aqueous solution was extracted with dichloromethane and the combined
organic layers were dried (MgSO4). Solvent was removed under reduced pressure to produce a brownish
oil. The crude product was purified by column chromatography on silica gel (EtOAc/n-heptane 7/3) to
provide pure benzophenone 3x as a white solid; 87% yield (0.52 g); mp (Et2O) 160-162 °C; 1H NMR (CDCl3,
400MHz) δ (ppm) 2.34 (s, 3H, OCOCH3), 3.89 (s, 6H, 2OCH3), 3.93 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 7.04 (s,
13
2H, ArH), 7.05 (d, J = 8.6 Hz, 1H, ArH), 7.58 (d, J = 2.0 Hz, 1H, ArH), 7.76 (dd, J = 8.6, 2.0 Hz, 1H, ArH). C
NMR (CDCl3, 100 MHz) δ 20.7 (CH3), 56.1 (CH3), 56.3 (2CH3), 61.0 (CH3), 107.4 (2CH), 111.5 (CH), 125.3 (CH),
129.8 (CH), 130.3 (C), 132.8 (C), 139.2 (C), 141.7 (C), 152.9 (2C), 154.8 (C), 168.9 (C), 193.8 (C). IR ν (cm-1):
1769, 1643, 1603, 1579, 1501, 1411, 1321, 1271, 1205, 1113, 1012, 910, 869, 760, 638.
4.2. Human pyruvate dehydrogenase kinase (PDHK1, 2 and 4) enzymatic radiometric assay [km ATP],
KinaseProfiler
PDHK1, 2 or 4 (h) is incubated with 8 mM MOPS pH 7.0, 0.2 mM EDTA, 250 uM KKKYHGHSMSDPGVSYRT,
10 mM Mg Acetate and [gamma-33P-ATP] (specific activity and concentration as required). The reaction is
initiated by the addition of the Mg/ATP mix. After incubation for 40 minutes at room temperature, the
21

reaction is stopped by the addition of phosphoric acid to a concentration of 0.5%. 10 μL of the reaction is
then spotted onto a P30 filtermat and washed four times for 4 minutes in 0.425% phosphoric acid and once
in methanol prior to drying and scintillation counting. (source: human recombinant; substrate: 250 μM
KKKYHGHSMSDPG; incubation: 40 min at rt; tracer: 33P; detection method: radiometric; measured response:
scintillation).
4.3. Tubulin polymerization assay
Sheep brain tubulin was purified according to the method of Shelanski [21] by two cycles assembly–
disassembly and then dissolved in the assembly buffer containing 0.1 M MES, 0.5 mM MgCl2, 1 mM EGTA,
and 1 mM of GTP (pH 6.6) to give a tubulin concentration of about 2–3 mg/mL. Tubulin assembly was
monitored by fluorescence according to reported procedure [22] using DAPI as fluorescent molecule.
Assays were realized on 96-well plates prepared with Biomek NKMC and Biomek 3000 from Beckman
coulter and read at 37 °C on Wallac Victor fluorimeter from Perkin–Elmer. The IC50 value of each compound
was determined as tubulin polymerization inhibition by 50% compared to the rate in the absence of
compound. The IC50 values for all compounds were compared to the IC50 of phenstatin and
desoxypodophyllotoxin and measured the same day under the same conditions.
4.4. Analysis of cell cycle
The impact of selected molecules on cell cycle of murine leukemia DA1-3b cells was evaluated after fixation
overnight at 4°C with 70% ice-cold ethanol/phosphate-buffered saline (PBS) followed by staining with a
solution containing PI (50 g/mL) and RNase A (0.5 mg/mL), and analyzed with a Becton Dickinson FACScan
cytofluorometer. DA1-3b cells were incubated for 24 h in presence of chloroacetyl-thiophenstatin 4a and
thiophenstatin 4b.
4.5. Cell proliferation assay
The compounds were tested against a panel of 60 human cancer cell lines at the National Cancer Institute,
Germantown, MD, USA [23]. The cytotoxicity studies were conducted using a 48 h exposure protocol using
the sulforhodamine B assay [24].
Acknowledgments
The authors gratefully acknowledge the National Cancer Institute (NCI) for the biological evaluation of
compounds on their 60-cell panel: the testing was performed by the Developmental Therapeutics Program,
Division of Cancer Treatment and Diagnosis (the URL to the Program’s website: http://dtp.cancer.gov).
Conflict of interest
The authors state no conflict of interest.
Supplementary data
Full description and graphs from NCI 60 cell one-dose and 5-dose screen and results of cell cycle distribution
for dichloroacetic acid (DCA) on DA1-3b cells are available in the ‘Supplementary information’ section.
References
[1] https://www.who.int/news-room/fact-sheets/detail/cancer, consulted on May 2019.
22

[2] M. P. Smolinski, Y. Bu, J. Clements, I. H. Gelman, T. Hegab, D. L. Cutler, J. W. S. Fang, G. Fetterly, R. Kwan,
A. Barnett, J. Y. N. Lau, D. G. Hangauer, J. Med. Chem. 61 (2018) 4704-4719.
[3] http://clinicaltrials.gov/, consulted on September 2019.
[4] G. Sutendra, E. D. Michelakis, Front. Oncol. 3 (2013) 1-11.
[5] C. Trapella, R. Voltan, E. Melloni, V. Tisato, C. Celeghini, S. Bianco, A. Fantinati, S. Salvadori, R. Guerrini,
P. Secchiero, G. Zauli, J. Med. Chem. 59 (2016) 59, 147-156.
[6] S.-L. Zhang, Z. Yang, X. Hu, K. Y. Tam, Bioorg. Med. Chem. Lett. 28 (2018) 3441-3445.
[7] A. Ghinet, B. Rigo, J.-P. Hénichart, D. Le Broc-Ryckewaert, J. Pommery, N. Pommery, X. Thuru, B. Quesnel,
P. Gautret, Bioorg. Med. Chem. 19 (2011) 6042-6054.
[8] A. Ghinet, A. Tourteau, B. Rigo, V. Stocker, M. Leman, A. Farce, J. Dubois, P. Gautret, Bioorg. Med. Chem.
21 (2013) 2932-2940.
[9] A. Ghinet, B. Rigo, P. Gautret, Rev. Roum. Chim. 57 (2012) 1065-1072.
[10] T. J. Curphey, J. Org. Chem. 67 (2002) 6461-6473.
[11] J.-P. Liou, J.-Y. Chang, C.-W. Chang, C.-Y. Chang, N. Mahindroo, F.-M. Kuo, H.-P. Hsieh, J. Med. Chem.
47 (2004) 2897-2905.
[12] A. Ghinet, P. Gautret, N. Van Hijfte, B. Lédé, J.-P. Hénichart, E. Bîcu, B. Rigo, U. Darbost, A. Daich, Chem.-
Eur. J. 20 (2014) 10117–10130.
[13] G. R. Pettit, B. Toki, D. L. Herald, P. Verdier-Pinard, M. R. Boyd, E. Hamel, R. K. Pettit, J. Med. Chem. 41 (1998)
1688-1695.
[14] N. Makita, J. Ishiguro, K. Suzuki, F. Nara, J. Pharm. Pharmacol. 69 (2017) 43-51.
[15] R. Vereecque, A. Saudemont, B. Quesnel, J. Gene Med. 6 (2004), 751-759.
[16] A. Ghinet, I.-M. Moise, B. Rigo, G. Homerin, A. Farce, J. Dubois, E. Bîcu, Bioorg. Med. Chem. 24 (2016)
2307-2317.
[17] M. Kato, J. Li, J. L. Chuang, D. Y. Chuang, Structure 15 (2007) 992-1004.
[18] H. Cho, I. Shin, K. Cho, H. Yoon, E. K. Yoo, M.-J. Kim, S. Park, I.-K. Lee, N. D. Kim, T. Sim, J. Med. Chem.
62 (2019) 8461-8479.
[19] B. L. Woolbright, D. Choudhary, A. Mikhalyuk, C. Trammel, S. Shanmugam, E. Abbott, C. C. Pilbeam, J.
A. Taylor, Mol. Cancer Ther. 17 (2018) 2004-2012.
[20] T. Liu, H. Yin, Oncol. Rep. 37 (2017) 193-200.
[21] M.L. Shelanski, F. Gaskin, C.R. Cantor, Proc. Natl. Acad. Sci. U. S. A. 70 (1973) 765-768.
[22] D. M. Barron, S. K. Chatterjee, R. Ravindra, ,R. Roof, E. Baloglu, D.G. Kingston, S. Bane, Anal. Biochem.
315 (2003) 49-56.
[23] Boyd, R. B. (1997) The NCI in vitro Anticancer Drug Discovery Screen. In Anticancer Drug Development
Guide; Preclinical Screening, Clinical Trials, and Approval; Teicher, B., Ed.; Humana Press Inc. Totowa: NJ,
pp 23–42.
[24] R. H. Shoemaker, Nature Rev. 6 (2006) 813-823.
23

 34 synthesized benzo(thio)phenones
 Inhibition of tubulin polymerization
 Inhibition of PDHK1
 Molecular docking on studied targets
 High cytotoxicity effect on NCI-60 cancer cell
line panel (GI50 <10 nM)
 Cell cycle analysis on DA1-3b cells
cancer cell death

Highlights
 34 synthesized benzo(thio)phenones
 Inhibition of tubulin polymerization
 Inhibition of PDHK1
 Unprecedented dual tubulin/PDHK1 inhibitors identified
 Molecular docking on studied targets
 Cell cycle analysis on DA1-3b cells
 High cytotoxicity effect on NCI-60 cancer cell line panel (GI50 <10 nM)
24

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered as
potential competing interests:
25